Paper Title
Automated Video Transcription and Summarisation

Abstract
The rapid growth of multimedia content on platforms like YouTube has spurred the need for automated systems capable of processing and summarising video data. Manual transcription and summarization are both time-consuming and labor-intensive, and often impractical at scale. This paper explores a novel framework for automated video transcription and summarisation using the YouTube Transcript API, Natural Language Toolkit (NLTK), and Hugging Face Transformers. The transcription process retrieves subtitles from YouTube videos, and two summarisation approaches are implemented: extractive summarisation using NLTK and abstractive summarisation using Hugging Face Transformers. The paper conducts a comparative analysis between the two methods, evaluating performance based on key metrics such as coherence, conciseness, and semantic relevance. The results indicate that Transformer-based summarisation produces more coherent, concise, and semantically rich summaries compared to traditional extractive methods, highlighting the promise of deep learning models for large-scale video content summarisation.